v2.1.233
Added 4
- Add GitLab merge request URL support to the --worktree flag and the claude agents view where MRs display as !N
- Add opt-in forward_user_identity apps gateway setting on Anthropic upstreams that sends the signed-in user's identity as headers for per-user spend attribution
- Add opt-in memory cgroup support for Bash tool commands on Linux via CLAUDE_CODE_TOOL_MEMORY_LIMIT environment variable to prevent runaway builds from stalling sessions
- Add CLAUDE_CODE_WEBFETCH_CACHE_TTL_MS environment variable to configure the WebFetch session URL cache TTL with a default of 15 minutes
Changed 6
- Improve claude self-hosted-runner session start time by creating the session branch without rewriting the working tree and eliminating two server round trips that blocked the agent's launch
- Improve apps gateway error forwarding so 400/413 errors from Vertex, Foundry, and Claude Platform on AWS upstreams carry the upstream's own message and fix a bug with auto-compact on apps gateway
- Improve claude plugin validate to check a bare .claude/skills directory and report SKILL.md files whose frontmatter fails to parse
- Improve screen reader mode so the /effort selector renders as a numbered list with a typed-number prompt and hint and dialog text is no longer clipped
- Improve print mode diagnostics by writing a [claude-code:unrecognized_model] line to stderr when a request goes out for a model ID Claude Code doesn't recognize and mapping it with modelOverrides to silence
- Change the GitHub app setup tip to no longer appear in repositories whose origin remote is on gitlab.com or bitbucket.org and make the enterprise marketplace tip cover non-GitHub internal git hosts
Fixed 9
- Fix cloud sessions occasionally being marked as lost when the environment shut down while Claude was waiting on a permission prompt
- Fix MCP v2 connections endlessly reopening the subscriptions/listen stream against servers that terminate long-held streams on a fixed timeout
- Fix Notification hooks not firing for permission prompts when running under Claude Desktop or VS Code
- Fix idle sessions on Linux sometimes keeping one CPU core at 100% when sandboxing is enabled
- Fix bundled skill aliases like /checkup and /review reporting Unknown command in -p mode or with plugins/MCP loaded when a user or project skill shadows the bundled skill
- Fix skill/command argument substitution to prevent argument values from being re-expanded as template markers
- Fix Windows paths spelled with the NT \??\ device prefix bypassing UNC path validation, closing an NTLM credential-leak vector
- Fix Windows auto mode repeatedly stopping for manual approval on ordinary cd <dir> && <command> > file Bash commands
- Revert the 2.1.232 Bash permission changes for Cygwin-style symlinks on Windows and for input redirections < file
Removed 1
- Remove todo/task-tracking tools (TaskCreate/Get/Update/List, TodoWrite) on Opus 4.8, Sonnet 5, Fable 5, Mythos 5, and newer models unless CLAUDE_CODE_ENABLE_TODO_TOOLS=1 is set
What's changed
- Added GitLab merge request URL support to the
--worktreeflag and theclaude agentsview (where MRs display as!N) - Added an opt-in
forward_user_identityapps gateway setting on Anthropic upstreams that sends the signed-in user's identity as headers, so a proxy behind the gateway can attribute spend per user - Added opt-in memory cgroup support for Bash tool commands on Linux (
CLAUDE_CODE_TOOL_MEMORY_LIMIT) so a runaway build can't stall the session - Added
CLAUDE_CODE_WEBFETCH_CACHE_TTL_MSenvironment variable to configure the WebFetch session URL cache TTL (default unchanged: 15 minutes) - Fixed cloud sessions occasionally being marked as lost when the environment shut down while Claude was waiting on a permission prompt
- Fixed MCP v2 connections endlessly reopening the subscriptions/listen stream against servers that terminate long-held streams on a fixed timeout (e.g. serverless hosts)
- Fixed Notification hooks not firing for permission prompts when running under Claude Desktop or VS Code
- Fixed idle sessions on Linux sometimes keeping one CPU core at 100% when sandboxing is enabled
- Fixed bundled skill aliases like
/checkupand/reviewreporting "Unknown command" in-pmode or with plugins/MCP loaded when a user or project skill shadows the bundled skill - Fixed skill/command argument substitution to prevent argument values from being re-expanded as template markers
- Fixed Windows paths spelled with the NT
\??\device prefix bypassing UNC path validation, closing an NTLM credential-leak vector - Improved
claude self-hosted-runnersession start time: the session branch is now created without rewriting the working tree, and two server round trips no longer block the agent's launch - Improved apps gateway error forwarding: 400/413 errors from Vertex, Foundry, and Claude Platform on AWS upstreams now carry the upstream's own message; fixes a bug with auto-compact on apps gateway
- Improved
claude plugin validateto check a bare.claude/skillsdirectory, reporting SKILL.md files whose frontmatter fails to parse - Improved screen reader mode: the
/effortselector renders as a numbered list with a typed-number prompt, and hint and dialog text is no longer clipped - Improved print mode diagnostics: a
[claude-code:unrecognized_model]line is written to stderr when a request goes out for a model ID Claude Code doesn't recognize; map it withmodelOverridesto silence - Changed the GitHub app setup tip to no longer appear in repositories whose origin remote is on gitlab.com or bitbucket.org; the enterprise marketplace tip now covers non-GitHub internal git hosts
- Todo/task-tracking tools (TaskCreate/Get/Update/List, TodoWrite) are no longer available on Opus 4.8, Sonnet 5, Fable 5, Mythos 5, and newer models; set
CLAUDE_CODE_ENABLE_TODO_TOOLS=1to bring them back - Windows: fixed auto mode repeatedly stopping for manual approval on ordinary
cd <dir> && <command> > fileBash commands (a 2.1.232 regression) - Reverted the 2.1.232 Bash permission changes for Cygwin-style symlinks on Windows and for input redirections (
< file); a narrower version will return in a later release